This is the mail archive of the
mailing list for the Cygwin project.
RE: [RFD]: Changing datatypes for uid_t/gid_t
- From: Mark Bradshaw <bradshaw at staff dot crosswalk dot com>
- To: 'Corinna Vinschen' <cygwin at cygwin dot com>
- Date: Mon, 21 Jan 2002 15:29:03 -0500
- Subject: RE: [RFD]: Changing datatypes for uid_t/gid_t
I'd vote for 32 bit to give maximum compatibility with other systems and old
> -----Original Message-----
> From: Corinna Vinschen [mailto:email@example.com]
> Sent: Monday, January 21, 2002 1:27 PM
> To: cygwin
> Subject: [RFD]: Changing datatypes for uid_t/gid_t
> due to the problems with uids and gids beyond 65535, we're thinking
> about changing the sizeof these datatypes from the current 16 bit
> to a bigger size. We discussed that already internally and we have
> a way to do it in a backward compatible way. Newer applications
> using the new data sizes would not run on older Cygwin DLLs but that's
> already true for many newer applications today so that shouldn't
> matter (at least I'd still sleep well).
> Already existing applications compiled using 16 bit uids/gids would
> still run, provided that you're only using 16 bit clean uids/gids
> with that applications. If /etc/passwd or /etc/group contain bigger
> uids/gids then... shit happens.
> We have three possible data sizes:
> - 16 bit. Don't touch anything. Advantage: Nothing to do for us,
> no compatibility problems at all.
> - 32 bit. Coincidentally equal to sizeof int and sizeof long.
> Advantage: Should even work with older software which often
> silently assume that a uid/gid can be stored in a int.
> - 64 bit. Advantage: Since the RIDs used in Windows are just
> 32 bit, one could create uids/gids with using the upper 32 bit
> of the uid/gid as a domain/server identifier, e.g.
> So, please tell us what you think is the better approach from your
> point of view. We're trying to get a picture.
> Corinna Vinschen Please, send mails
> regarding Cygwin to
> Cygwin Developer
> Red Hat, Inc.
> Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
> Bug reporting: http://cygwin.com/bugs.html
> Documentation: http://cygwin.com/docs.html
> FAQ: http://cygwin.com/faq/
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html